Source return exception processing method and device and computing equipment

Abstract

The invention discloses a back-to-source exception processing method and device and computing equipment. The method comprises the following steps: after any node of the content distribution network initiates a source return request, generating a source return record corresponding to the source return request; the back-to-source record comprises cluster information of a cluster to which the back-to-source request initiating node belongs, resource information of resources corresponding to the back-to-source request, and first node information of the back-to-source request processing node; obtaining a back-to-source record generated in a preset time window; detecting repeatedly appearing target back-to-source records from the back-to-source records; and according to the detected target back-to-source record, determining whether abnormal repeated back-to-source exists in a preset time window or not. According to the scheme, whether the abnormal repeated source return occurs in the preset time window can be accurately detected according to the source return record, rapid and accurate detection of the abnormal repeated source return is realized, the source return anomaly detection efficiency and precision are improved, the detection cost is saved, and the method can adapt to different protocols.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a method, apparatus, computing device, computer storage medium, and computer program product for back-to-source exception processing. Background Art

[0002] Content Delivery Network (CDN) is a network that places service nodes in multiple locations, with one or more service nodes processing user requests. In application scenarios such as live broadcasting, the content delivery network will go back to the source when processing user requests, that is, transmit the resources requested by the user to the user end through one or more nodes in the content delivery network.

[0003] Content distribution networks usually perform request convergence when performing back-to-source, thereby converging multiple back-to-source requests into one back-to-source path. However, if there are anomalies in the back-to-source algorithm in the content distribution network, there will be anomalies in request convergence. The existing technology usually adopts the method of manual analysis of anomalies, but this method is inefficient and costly. Summary of the invention

[0004] In view of the above problems, the present application is proposed to provide a method, apparatus, computing device, computer storage medium and computer program product for handling back-to-source exceptions that overcome the above problems or at least partially solve the above problems.

[0005] According to the first aspect of the present application, a method for handling back-to-source exceptions is provided, comprising:

[0006] After any node of the content distribution network initiates a back-to-source request, a back-to-source record corresponding to the back-to-source request is generated; wherein the back-to-source record includes cluster information of the cluster to which the node initiating the back-to-source request belongs, resource information of the resource corresponding to the back-to-source request, and first node information of the node processing the back-to-source request;

[0007] Get the back-to-source records generated within the preset time window;

[0008] Detecting repeated target back-to-source records from the back-to-source records;

[0009] Determine whether there is abnormal repeated return to the source within the preset time window based on the detected target return to the source record.

[0010] In an optional implementation, the method further includes: determining the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record;

[0011] Then, determining whether there is abnormal repeated return to the source within the preset time window according to the detected target return to the source record includes: determining whether there is abnormal repeated return to the source within the preset time window according to the number of target return to the source records and / or the occurrence frequency of any target return to the source record.

[0012] In an optional implementation, determining whether there is abnormal repeated back-to-source within a preset time window according to the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record includes:

[0013] If the number of target back-to-source records exceeds a preset number threshold, it is determined that abnormal duplicate back-to-source exists within the preset time window;

[0014] And / or, according to the occurrence frequency of each target back-to-source record, the occurrence frequency accumulation value of the target back-to-source record in the preset time window is calculated, and if the occurrence frequency accumulation value exceeds a preset accumulation threshold, it is determined that abnormal repeated back-to-source exists in the preset time window.

[0015] In an optional implementation, the method further includes: if it is determined that abnormal repeated return to the source exists within a preset time window, generating an alarm message.

[0016] In an optional embodiment, the method further includes:

[0017] If it is determined that there is abnormal repeated return to the source within the preset time window;

[0018] Extract cluster information from each target back-to-source record;

[0019] For any extracted cluster information, determine the duplication index of the cluster information according to the occurrence frequency of the target back-to-source record corresponding to the cluster information;

[0020] Generate matching alarm information based on the duplication indicators of each cluster information.

[0021] In an optional implementation, generating matching alarm information according to the duplication index of each cluster information includes:

[0022] Add the repeated indicators of each cluster information to the repeated indicator set;

[0023] Calculating a volatility index of the set of repeated indicators;

[0024] If the fluctuation index is greater than the preset fluctuation index threshold, the target cluster information is determined from the cluster information according to the repetition index of each cluster information, and an alarm message is generated indicating that the back-to-source algorithm of the target cluster corresponding to the target cluster information is abnormal;

[0025] If the fluctuation index is less than or equal to the preset fluctuation index threshold, an alarm message indicating an abnormal cross-cluster backhaul algorithm is generated.

[0026] In an optional implementation, after generating the back-to-source record corresponding to the back-to-source request, the method further includes: recording second node information of the back-to-source request initiating node corresponding to each back-to-source record;

[0027] After acquiring the back-to-source record generated within the preset time window, the method further includes: merging multiple back-to-source records corresponding to the same second node information into one back-to-source record;

[0028] And / or, if the number of source records corresponding to the same second node information exceeds a preset record number threshold, an alarm message indicating that the node corresponding to the second node information is abnormal is generated.

[0029] According to a second aspect of the present application, a back-to-source exception handling device is provided, including:

[0030] A generation module, configured to generate a back-to-source record corresponding to a back-to-source request after any node of a content distribution network initiates a back-to-source request; wherein the back-to-source record includes cluster information of a cluster to which the node initiating the back-to-source request belongs, resource information of a resource corresponding to the back-to-source request, and first node information of a node processing the back-to-source request;

[0031] The acquisition module is used to obtain the back-to-source records generated within a preset time window;

[0032] A detection module, used to detect repeated target back-to-source records from the back-to-source records;

[0033] The determination module is used to determine whether there is abnormal repeated return to the source within a preset time window according to the detected target return to the source record.

[0034] In an optional implementation, the detection module is further used to: determine the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record;

[0035] The determination module is used to determine whether there is abnormal repeated return to the source within a preset time window according to the number of target return to the source records and / or the occurrence frequency of any target return to the source record.

[0036] In an optional implementation, the determination module is used to: if the number of target back-to-source records exceeds a preset number threshold, determine that abnormal repeated back-to-source exists within a preset time window;

[0037] And / or, according to the occurrence frequency of each target back-to-source record, the occurrence frequency accumulation value of the target back-to-source record in the preset time window is calculated, and if the occurrence frequency accumulation value exceeds a preset accumulation threshold, it is determined that abnormal repeated back-to-source exists in the preset time window.

[0038] In an optional implementation, the device further includes an alarm module configured to generate an alarm message if it is determined that abnormal repeated return to the source exists within a preset time window.

[0039] In an optional implementation, the alarm module is used to: if it is determined that there is abnormal repeated return to the source within a preset time window;

[0040] Extract cluster information from each target back-to-source record;

[0041] For any extracted cluster information, determine the duplication index of the cluster information according to the occurrence frequency of the target back-to-source record corresponding to the cluster information;

[0042] Generate matching alarm information based on the duplication indicators of each cluster information.

[0043] In an optional implementation, the alarm module is used to: add the repetition index of each cluster information to the repetition index set;

[0044] Calculating a volatility index of the set of repeated indicators;

[0045] If the fluctuation index is greater than the preset fluctuation index threshold, the target cluster information is determined from the cluster information according to the repetition index of each cluster information, and an alarm message is generated indicating that the back-to-source algorithm of the target cluster corresponding to the target cluster information is abnormal;

[0046] If the fluctuation index is less than or equal to the preset fluctuation index threshold, an alarm message indicating an abnormal cross-cluster backhaul algorithm is generated.

[0047] In an optional implementation, the device further includes: a recording module, configured to record second node information of a node initiating the back-to-source request corresponding to each back-to-source record after generating the back-to-source record corresponding to the back-to-source request;

[0048] The acquisition module is used to: after acquiring the back-to-source record generated within the preset time window, merge multiple back-to-source records corresponding to the same second node information into one back-to-source record;

[0049] And / or, an alarm module, for generating an alarm message indicating that a node corresponding to the second node information is abnormal if the number of source records corresponding to the same second node information exceeds a preset record number threshold.

[0050] According to a third aspect of the present application, there is provided a computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0051] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned return-to-source exception handling method.

[0052] According to a fourth aspect of the present application, a computer storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned back-to-source exception handling method.

[0053] According to a fifth aspect of the present application, a computer program product is provided, comprising at least one executable instruction, wherein the executable instruction enables a processor to perform operations corresponding to the above-mentioned back-to-source exception handling method.

[0054] After initiating a back-to-source request, the embodiment of the present application generates a back-to-source record containing relevant information of the cluster where the initiating node is located, the requested resources, and the next-hop node, so that according to the back-to-source record, it is possible to accurately detect whether an abnormal repeated back-to-source occurs within a preset time window, thereby achieving rapid and accurate detection of abnormal repeated back-to-source, improving the efficiency and accuracy of back-to-source anomaly detection, saving detection costs, and being adaptable to different protocols.

[0055] The embodiment of the present application determines whether there is abnormal repeated return to the source according to the number and / or occurrence frequency of the target return to the source record, thereby improving the accuracy of determining the abnormal repeated return to the source.

[0056] The embodiment of the present application generates an alarm message after determining that an abnormal repeated return to the source exists within a preset time window according to the detection result, thereby facilitating timely abnormal processing of the alarm message.

[0057] In the embodiment of the present application, when it is determined that abnormal repeated back-to-source occurs in a preset time window, the cause of the abnormality is further located based on the cluster information of the target back-to-source record, and matching alarm information is generated according to the determined abnormal cause, so as to facilitate the rapid determination of the abnormal cause according to the alarm information and the rapid processing of the abnormal point; moreover, in the process of locating the cause of the abnormality, the repetition index of the cluster information in the target back-to-source record is calculated, the cause of the abnormality is determined according to the repetition index, and matching alarm information is generated, so as to improve the accurate positioning of the abnormal cause of the abnormal repeated back-to-source.

[0058] The embodiment of the present application can accurately determine whether the abnormal repeated backtracking is caused by an abnormal backtracking algorithm within a cluster or an abnormal backtracking algorithm across clusters based on the fluctuation index of the repeated indicator set, thereby improving the accuracy of determining the cause of the abnormal repeated backtracking.

[0059] In the back-to-source exception processing method provided in the embodiment of the present application, after obtaining the back-to-source record generated within the preset time window, multiple back-to-source records corresponding to the same second node information are merged into one back-to-source record, thereby avoiding the interference of the initiating node anomaly on the abnormal repeated back-to-source result, and avoiding the interference of the initiating node anomaly on the subsequent intra-cluster or cross-cluster back-to-source algorithm abnormal judgment result, thereby improving the detection accuracy of abnormal repeated back-to-source and the accuracy of determining the cause of the anomaly; and the number of records of the back-to-source record corresponding to the second node information can quickly determine whether the initiating node has an abnormality, which is convenient for locating the abnormality of a single CDN node.

[0060] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0062] Figure 1 An operation architecture diagram provided by Embodiment 1 of the present application is shown;

[0063] Figure 2 A schematic diagram of a flow chart of a method for handling a back-to-source exception provided in Embodiment 2 of the present application is shown;

[0064] Figure 3 A schematic diagram of a flow chart of a method for handling a back-to-source exception provided in Embodiment 3 of the present application is shown;

[0065] Figure 4 A schematic diagram of a flow chart of a method for handling a back-to-source exception provided in Embodiment 4 of the present application is shown;

[0066] Figure 5 A schematic diagram of the structure of a back-to-source exception handling device provided in Embodiment 5 of the present application is shown;

[0067] Figure 6 A structural schematic diagram of a computing device provided in Example 6 of the present application is shown. DETAILED DESCRIPTION

[0068] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0069] Embodiment 1

[0070] Figure 1 An operation architecture diagram provided by the first embodiment of the present application is shown. The operation architecture diagram is applied to a client and a content distribution network.

[0071] The client can be running Windows, Android TM ) or IOS and other operating systems, such as smart phones, tablet devices, laptops, virtual reality devices, gaming devices, set-top boxes, car terminals, smart TVs. Based on the above operating systems, various applications can be run, such as browsers, live broadcast software, etc.

[0072] The content distribution network includes at least one CDN node and an origin station, and multiple CDN nodes form corresponding clusters according to their geographical locations, etc. The cluster refers to a set of nodes that are interrelated and work together to complete a task. Usually, the CDN node close to the client side is called an edge node, the node that provides and stores the original data source in the content distribution network is called an origin station, and the node between the origin station and the edge node is called an intermediate node. The nodes in the content distribution network can be composed of a single or multiple computing devices. The computing device may include a virtualized computing instance. The virtualized computing instance may include a virtual machine, such as a simulation of a computer system, an operating system, a server, etc. The computing device may load a virtual machine based on a virtual image and / or other data that defines specific software (e.g., an operating system, a dedicated application, a server) for simulation. As the demand for different types of processing services changes, different virtual machines may be loaded and / or terminated on one or more computing devices. A hypervisor may be implemented to manage the use of different virtual machines on the same computing device.

[0073] Nodes in a content delivery network may be configured to communicate with clients or other nodes via a network. The network includes various network devices, such as routers, switches, multiplexers, hubs, modems, bridges, repeaters, firewalls, proxy devices, and / or the like. The network may include physical links, such as coaxial cable links, twisted pair cable links, optical fiber links, combinations thereof, and the like, or wireless links, such as cellular links, satellite links, Wi-Fi links, and the like.

[0074] Embodiment 2

[0075] Figure 2 A flow chart of a method for handling back-to-source exceptions provided in Embodiment 2 of the present application is shown.

[0076] The method can be applied to a preset device in a content distribution network.

[0077] Specifically, Figure 2 As shown, the method comprises the following steps:

[0078] Step S210, after any node of the content distribution network initiates a back-to-source request, a back-to-source record corresponding to the back-to-source request is generated; wherein the back-to-source record includes cluster information of the cluster to which the node initiating the back-to-source request belongs, resource information of the resource corresponding to the back-to-source request, and first node information of the node processing the back-to-source request.

[0079] Other nodes other than the source station in the content distribution network (such as edge nodes and / or intermediate nodes, etc.) will request back-to-source from the next node in the content distribution network when their own cached data cannot meet the client's needs, and the node will initiate a back-to-source request.

[0080] After any node of the content distribution network initiates a return-to-source request, a return-to-source record corresponding to the return-to-source request is generated, and the return-to-source record corresponds to the return-to-source request one by one. After the return-to-source record corresponding to the return-to-source request is generated, the generated return-to-source record is stored. Among them, the embodiment of the present application does not limit the recording method, storage method, and execution subject of the return-to-source record. For example, the return-to-source record can be recorded in the log file of the node initiating the return-to-source request in the form of a preset field, or a separate record data can be generated for storage, and the generated return-to-source record can be submitted to the execution device of this method in a timely manner. In addition, the return-to-source record can be generated by the execution device of this method according to the analysis of the return-to-source request, and can also be executed by the return-to-source layer of the node initiating the return-to-source request. Since the return-to-source request is initiated by the return-to-source layer of the node initiating the return-to-source request, the generation of the return-to-source record by the return-to-source layer can improve the generation accuracy and generation efficiency of the return-to-source record, so the embodiment of the present application can preferably generate the corresponding return-to-source record by the return-to-source layer of the node initiating the return-to-source request.

[0081] In an optional implementation, a back-to-source record can be generated by parsing a back-to-source request. Specifically, different back-to-source requests can use the same or different protocols (such as a custom protocol, hlsv7 protocol, etc.). In order to improve the detection accuracy of subsequent target back-to-source records and avoid the impact of different protocols on the detection results, the back-to-source records generated by different back-to-source requests in this implementation are consistent in record format. In the specific implementation process, a matching extraction algorithm (such as a regular matching algorithm or a semantic recognition algorithm, etc.) can be generated for different protocols, and the extraction algorithm is used to extract the corresponding information from the back-to-source request of the corresponding protocol, and the back-to-source record is generated according to the preset format based on the information.

[0082] Specifically, the back-to-source information corresponding to the back-to-source request contains the cluster information of the cluster to which the node initiating the back-to-source request belongs. The node initiating the back-to-source request is the CDN node that initiates the back-to-source request, and the cluster where the CDN node is located is the cluster to which the node initiating the back-to-source request belongs. The back-to-source information records the cluster information of the cluster, which is information that can uniquely identify the cluster, such as the cluster name, cluster ID, cluster address, etc.

[0083] The back-to-source information corresponding to the back-to-source request contains the resource information of the resource corresponding to the back-to-source request. The purpose of the CDN node initiating the back-to-source request is to obtain the corresponding resources from other CDN nodes, such as obtaining live streaming data, etc. The resource requested by the back-to-source request is the resource corresponding to the back-to-source request. The back-to-source information records the resource information of the resource, which is the information that can uniquely identify the resource, such as the resource name, resource ID, etc.

[0084] The back-to-source information corresponding to the back-to-source request contains the first node information of the back-to-source request processing node. The back-to-source request initiating node initiates the back-to-source request, and the node where the back-to-source request reaches and processes the back-to-source request is the back-to-source request processing node, which can also be called the next hop node of the back-to-source request initiating node. The back-to-source information records the node information of the back-to-source request node, which is the first node information, and the first node information can include a node identifier (such as a node name or node ID), an IP address, and the like.

[0085] The cluster information, resource information, and first node information in the back-to-source record are arranged in a preset format. For example, the back-to-source record can be represented as unique_url, idc represents cluster information, remoteip represents first node information, and stream represents resource information. Then unique_url can be represented as: idc_stream_remoteip. For example, the cluster name of the CDN node N1 in the content distribution network is "Sss". The CDN node N1 initiates a back-to-source request to the CDN node with the address "10.10.10.10", requesting to obtain the live streaming data with the stream name "live-111". After initiating the back-to-source request, the CDN node N1 will generate a back-to-source record "Sss_live-111_10.10.10.10". The back-to-source record can be generated by the back-to-source layer of the CDN node N1 and recorded in its log file, and the back-to-source record can be identified with a preset identifier. For example, whether it is a long link (such as an flv request) or a short link (such as an hlsv7 request), or other protocols, after the back-to-source record is generated, the back-to-source record is marked with error=9999.

[0086] In addition, each generated back-to-source record is assigned a timestamp according to its generation time, so that the generation time of the back-to-source record can be determined through the timestamp of the back-to-source record.

[0087] Step S220: Obtain the back-to-source records generated within a preset time window.

[0088] From the back-to-source records corresponding to each node in the generated content distribution network, filter out the back-to-source records whose generation time is within the preset time window. The preset time window is a historical time window, such as a time window whose end point is the current time. The preset time window can be the last 5 minutes, the last 10 minutes, etc.

[0089] In an optional implementation, this step can be performed regularly, and the execution cycle can be consistent with the length of the preset time window. For example, this step is performed every 5 minutes, and the back-to-source records generated in the last 5 minutes are obtained each time the step is performed.

[0090] In an optional implementation, the window length of the preset time window can be dynamically adjusted according to actual business conditions. For example, the preset time window can be dynamically adjusted in one or more of the following ways.

[0091] Method 1: Dynamically adjust the length of the preset time window according to the status of the content distribution network update (such as the launch of a new scheduling algorithm, different version iterations, etc.). Specifically, set a benchmark preset time window, and in the first time period after the content distribution network update is detected (such as one hour, the first time period is greater than each preset time window), shorten the length of the benchmark preset time window to obtain the first preset time window, and filter the back-to-source records with the first preset time window (the first preset time window is used as the preset time window) in the first time period; filter the back-to-source records with the benchmark preset time window (the benchmark preset time window is used as the preset time window) in other time periods other than the first time period after the content distribution network is updated; if there are no new update events in the second time period (such as two hours), extend the benchmark preset time window to obtain the second preset time window, and filter the back-to-source records with the second preset time window (the second preset time window is used as the preset time window).

[0092] Method 2: Dynamically adjust the length of the next preset time window according to the number and / or frequency of occurrence of the target source return records in step S230 of the previous preset time window. Specifically, if the number and / or frequency of occurrence of the target source return records in the previous preset time window per unit time exceeds the preset numerical range, the window length of the previous preset time window is shortened by the preset length as the window length of the next preset time window. If the number and / or frequency of occurrence of the target source return records in the previous preset time window per unit time does not reach the preset numerical range, the window length of the previous preset time window is extended by the preset length as the window length of the next preset time window.

[0093] Step S230: Detect repeated target back-to-source records from the back-to-source records.

[0094] The target back-to-source record is detected from the back-to-source record obtained in step S220. The target back-to-source record is a back-to-source record that appears repeatedly in the back-to-source record obtained in step S220. Specifically, if multiple back-to-source records are the same, it means that the same resource is requested from the same cluster to the same node, so it is determined that there is an abnormality in the current request convergence and repeated back-to-source occurs, and the same multiple back-to-source records are determined as one target back-to-source record.

[0095] In an optional implementation, the number of target back-to-source records detected within a preset time window is determined, and / or, for any target back-to-source record, the occurrence frequency of the target back-to-source record in the back-to-source records acquired in step S220 is determined.

[0096] Step S240: Determine whether there is abnormal repeated back-to-source within a preset time window based on the detected target back-to-source record.

[0097] If there is a target source return record within the preset time window, it indicates that there is a repeated source return within the preset time window, so whether there is an abnormal repeated source return within the preset time window is determined according to the detected target source return record.

[0098] In an optional implementation, it is determined whether there is abnormal repeated return to the source within a preset time window based on the number of target return to the source records and / or the frequency of occurrence of any target return to the source record. Specifically, if the number of target return to the source records exceeds a preset number threshold, it is determined that there is abnormal repeated return to the source within the preset time window; and / or, based on the frequency of occurrence of each target return to the source record, the cumulative value of the frequency of occurrence of the target return to the source record within the preset time window is calculated (the cumulative value of the frequency of occurrence is the sum of the frequency of occurrence of each target return to the source record within the preset time window), and if the cumulative value of the frequency of occurrence exceeds a preset cumulative threshold, it is determined that there is abnormal repeated return to the source within the preset time window.

[0099] Further optionally, in this embodiment, a historical period is obtained (the historical period is N times the length of the preset time window, such as the preset time window is 5 minutes, then the historical period is one week, that is, N is a larger value than the preset value), and the number and / or the cumulative value of the target back-to-source records in the historical period are counted, and the preset number threshold in the preset time window is obtained according to the ratio of the number to N, and the preset cumulative threshold in the preset time window is obtained according to the ratio of the cumulative value of the frequency of occurrence to N. Therefore, the preset number threshold and the preset cumulative threshold in this embodiment are determined according to the average value in a longer historical period, and the preset number threshold and the preset cumulative threshold can reflect the frequency of repeated back-to-source under the average state, so that the abnormal repeated back-to-source is determined according to the preset number threshold and the preset cumulative threshold with a higher determination accuracy.

[0100] In an optional implementation, if it is determined that there is an abnormal repeated return to the source within the preset time window, a corresponding alarm message is generated. In addition, the alarm message can also be sent to a relevant terminal (such as a terminal device of a maintenance personnel, etc.) for timely processing of the abnormal repeated return to the source. Further optionally, the alarm message can also carry corresponding auxiliary information, such as the number of target return to the source records within the preset time window and / or the accumulated value of the frequency of occurrence, etc., so as to facilitate the rapid positioning of the abnormal point.

[0101] It can be seen that in the back-to-source exception handling method provided in the embodiment of the present application, after initiating a back-to-source request, a back-to-source record containing relevant information of the cluster where the initiating node is located, the requested resources, and the next-hop node is generated, so that according to the back-to-source record, it is possible to accurately detect whether an abnormal repeated back-to-source occurs within the preset time window, thereby achieving rapid and accurate detection of abnormal repeated back-to-source, improving the efficiency and accuracy of back-to-source anomaly detection, saving detection costs, and being able to adapt to different protocols.

[0102] Embodiment 3

[0103] Figure 3 A flow chart of a method for handling a back-to-source exception provided in Embodiment 3 of the present application is shown.

[0104] The method can be applied to a preset device in a content distribution network.

[0105] Specifically, Figure 3 As shown, the method comprises the following steps:

[0106] Step S310: If it is determined that there is abnormal repeated return to the source within the preset time window.

[0107] The method of determining whether there is abnormal repeated return to the source can refer to the description in other method embodiments, which will not be repeated here.

[0108] Step S320: extract cluster information from each target back-to-source record.

[0109] For each target back-to-source record, cluster information in the target back-to-source record is extracted, so as to obtain the cluster information appearing in the target back-to-source record in aggregating form.

[0110] Step S330 : for any extracted cluster information, determine the repetition index of the cluster information according to the occurrence frequency of the target back-to-source record corresponding to the cluster information.

[0111] For any cluster information extracted in step S320, determine the target back-to-source record corresponding to the cluster information, and determine the repetition index of the cluster information according to the occurrence frequency of the corresponding target back-to-source record. Specifically, the cumulative value of the occurrence frequency of each target back-to-source record corresponding to the cluster information is used as the repetition index of the cluster information. The higher the repetition index, the more serious the repeated back-to-source phenomenon of the cluster within the preset time period.

[0112] Step S340: Generate matching alarm information according to the duplication index of each cluster information.

[0113] According to the different distribution of duplicate indicators of each cluster information, different abnormal causes of duplicate back-to-source are located, so as to generate matching alarm information.

[0114] Specifically, the repetitive indicators of each cluster information are added to the repetitive indicator set, and the volatility index of the repetitive indicator set is calculated. The volatility index is used to express the volatility or non-uniformity of the set of data in the repetitive indicator set. The larger the volatility index, the more non-uniform the set of data of the repetitive indicator. Among them, this embodiment does not limit the specific determination method of the volatility index. For example, the variance, standard deviation, and interquartile deviation of the repetitive indicator set can be calculated, and the variance, standard deviation, or interquartile deviation can be used as the volatility index.

[0115] Further, according to the relationship between the volatility index and the preset volatility index threshold, it is determined whether the abnormality is caused by the intra-cluster retrieval algorithm or the cross-cluster retrieval algorithm. Optionally, historical data under different scenarios can be obtained respectively, and the scenarios include cross-cluster retrieval algorithm abnormality, single or multiple cluster retrieval algorithm abnormality, and the volatility index of the set of repeated indicators under different scenarios is determined according to the historical data under different scenarios, and then the preset volatility index threshold is determined comprehensively according to the volatility index under different scenarios. For example, the volatility index R1 under the cross-cluster retrieval algorithm abnormality, the volatility index R2 under the single cluster retrieval algorithm abnormality, and the volatility index R3 under the multiple cluster retrieval algorithm abnormality are counted, and it is determined that the preset volatility index threshold is less than R1, but greater than R2 and R3.

[0116] Specifically, if the fluctuation index is greater than the preset fluctuation index threshold, it indicates that the abnormal back-to-source frequency distribution of different clusters is uneven, indicating that there is an abnormality in the back-to-source algorithm within one or several clusters. In this case, the target cluster information is determined from the cluster information according to the repetition index of each cluster information, and an alarm message is generated that the back-to-source algorithm of the target cluster corresponding to the target cluster information is abnormal. For example, the cluster information can be sorted from high to low according to the repetition index, and the first N cluster information can be selected as the target cluster information, or the cluster information with a repetition index higher than the average value of the repetition index set can be selected as the target cluster information, and the cluster corresponding to the target cluster information is the target cluster. The screened target cluster has a large number of repeated back-to-source times, thereby generating an alarm message that the back-to-source algorithm of the target cluster is abnormal, which facilitates timely and accurate positioning of the abnormal point according to the alarm message.

[0117] Further optionally, if the fluctuation index is greater than a preset fluctuation index threshold, after identifying the target cluster information, the other cluster information other than the target cluster information contained in the target source return record is used as non-target cluster information, and the cluster corresponding to each non-target cluster information is used as a non-target cluster. Since the non-target cluster still corresponds to a certain number of target source return records with repeated source return, and the probability of abnormality in the source return algorithm within the non-target cluster is relatively small, the repeated source return phenomenon in the non-target cluster is usually affected by factors such as the online and offline nodes within the cluster. In this case, the present embodiment further determines the target source return record corresponding to each non-target cluster information, and for the target source return record corresponding to any non-target cluster information, determines the resource information contained in the target source return record, and calls the source return algorithm of the non-target cluster corresponding to the non-target cluster information to determine the source return request initiating node that currently matches the resource information, and subsequently the source return request initiating node initiates a source return request for the resource corresponding to the resource information. For example, cluster 1 is a non-target cluster, but there is a target back-to-source record corresponding to cluster 1, which appears twice, one time initiated by node N1 and the other time initiated by node N2. The resource information of the target back-to-source record is F0. The back-to-source scheduling algorithm in cluster 1 determines that the back-to-source request initiating node currently matching F0 is N2, then the resource corresponding to the resource information is redirected, and node N2 initiates the back-to-source request for the resource uniformly, directing the original back-to-source connection with node N1 to node N2.

[0118] If the fluctuation index is less than or equal to the preset fluctuation index threshold, it indicates that the abnormal backhaul frequencies of different clusters are evenly distributed, indicating that there is an abnormality in the cross-cluster backhaul algorithm between clusters (such as the cross-cluster backhaul scheduling algorithm, etc.). In this case, an alarm message of the cross-cluster backhaul algorithm abnormality is generated, so as to locate the abnormal point in a timely and accurate manner according to the alarm message.

[0119] It can be seen that the method for handling back-to-source exceptions provided in the embodiment of the present application, when it is determined that an abnormal repeated back-to-source occurs in a preset time window, further locates the cause of the exception based on the cluster information of the target back-to-source record, and generates matching alarm information according to the determined cause of the exception, thereby facilitating the rapid determination of the cause of the exception based on the alarm information and facilitating the rapid processing of the abnormal point; moreover, in the process of locating the cause of the exception, the repetition index of the cluster information in the target back-to-source record is calculated, the cause of the exception is determined based on the repetition index, and matching alarm information is generated, thereby improving the accurate positioning of the cause of the abnormal repeated back-to-source.

[0120] Embodiment 4

[0121] Figure 4 A flow chart of a method for handling a back-to-source exception provided in Embodiment 4 of the present application is shown.

[0122] The method can be applied to a preset device in a content distribution network.

[0123] Specifically, Figure 4 As shown, the method comprises the following steps:

[0124] Step S410: After any node of the content distribution network initiates a back-to-source request, a back-to-source record corresponding to the back-to-source request is generated, and the second node information of the node initiating the back-to-source request corresponding to each back-to-source record is recorded. The back-to-source record includes the cluster information of the cluster to which the node initiating the back-to-source request belongs, the resource information of the resource corresponding to the back-to-source request, and the first node information of the node processing the back-to-source request.

[0125] In the embodiment of the present application, in addition to recording the generated back-to-source records, the node information of the node initiating the back-to-source request of each back-to-source record is also recorded, and the node information is called the second node information. That is, the second node information is not part of the back-to-source record, but the second node information can be associated with the corresponding back-to-source record for storage.

[0126] Step S420: Acquire the back-to-source records generated within the preset time window, and merge multiple back-to-source records corresponding to the same second node information into one back-to-source record.

[0127] After obtaining the back-to-source records generated within the preset time window, the obtained back-to-source records are pre-processed. Specifically, in the obtained back-to-source records, multiple back-to-source records corresponding to the same second node information are merged into one back-to-source record. Subsequent processing is performed based on the merged back-to-source records.

[0128] In an optional implementation, if the number of back-to-source records corresponding to the same second node information exceeds a preset record number threshold, indicating that the CDN node corresponding to the second node information frequently initiates back-to-source requests within a preset time window, it is determined that the CDN node is abnormal, thereby generating an alarm message indicating that the node corresponding to the second node information is abnormal. The CDN node can also be subjected to network detection, restart, and other processing operations.

[0129] Step S430: Detect repeated target back-to-source records from the back-to-source records, and determine whether there is abnormal repeated back-to-source within a preset time window based on the detected target back-to-source records.

[0130] It can be seen that in the back-to-source exception processing method provided in the embodiment of the present application, after obtaining the back-to-source record generated within the preset time window, multiple back-to-source records corresponding to the same second node information are merged into one back-to-source record, thereby avoiding the interference of the initiating node anomaly on the abnormal repeated back-to-source result, and avoiding the interference of the initiating node anomaly on the subsequent intra-cluster or cross-cluster back-to-source algorithm abnormal judgment result, thereby improving the detection accuracy of abnormal repeated back-to-source and the accuracy of determining the cause of the anomaly; and the number of records of the back-to-source record corresponding to the second node information can quickly determine whether the initiating node has an abnormality, which is convenient for locating the abnormality of a single CDN node.

[0131] Embodiment 5

[0132] Figure 5 FIG. 5 shows a schematic diagram of the structure of a back-to-source exception handling device provided in Embodiment 5 of the present application. Figure 5 As shown, the device 500 includes: a generating module 510 , an acquiring module 520 , a detecting module 530 , and a determining module 540 .

[0133] A generation module 510 is used to generate a back-to-source record corresponding to a back-to-source request after any node of the content distribution network initiates a back-to-source request; wherein the back-to-source record includes cluster information of the cluster to which the node initiating the back-to-source request belongs, resource information of resources corresponding to the back-to-source request, and first node information of the node processing the back-to-source request;

[0134] The acquisition module 520 is used to acquire the back-to-source records generated within a preset time window;

[0135] A detection module 530, configured to detect repeated target back-to-source records from the back-to-source records;

[0136] The determination module 540 is used to determine whether there is abnormal repeated back-to-source within a preset time window according to the detected target back-to-source record.

[0137] In an optional implementation, the detection module 530 is further used to: determine the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record;

[0138] The determination module 540 is used to determine whether there is abnormal repeated back-to-source within a preset time window according to the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record.

[0139] In an optional implementation, the determination module 540 is used to: if the number of target back-to-source records exceeds a preset number threshold, determine that abnormal repeated back-to-source exists within a preset time window;

[0140] And / or, according to the occurrence frequency of each target back-to-source record, the occurrence frequency accumulation value of the target back-to-source record in the preset time window is calculated, and if the occurrence frequency accumulation value exceeds a preset accumulation threshold, it is determined that abnormal repeated back-to-source exists in the preset time window.

[0141] In an optional implementation, the device further includes an alarm module (not shown in the figure), which is configured to generate an alarm message if it is determined that abnormal repeated return to the source exists within a preset time window.

[0142] In an optional implementation, the alarm module is used to: if it is determined that there is abnormal repeated return to the source within a preset time window;

[0143] Extract cluster information from each target back-to-source record;

[0144] For any extracted cluster information, determine the duplication index of the cluster information according to the occurrence frequency of the target back-to-source record corresponding to the cluster information;

[0145] Generate matching alarm information based on the duplication indicators of each cluster information.

[0146] In an optional implementation, the alarm module is used to: add the repetition index of each cluster information to the repetition index set;

[0147] Calculating a volatility index of the set of repeated indicators;

[0148] If the fluctuation index is greater than the preset fluctuation index threshold, the target cluster information is determined from the cluster information according to the repetition index of each cluster information, and an alarm message is generated indicating that the back-to-source algorithm of the target cluster corresponding to the target cluster information is abnormal;

[0149] If the fluctuation index is less than or equal to the preset fluctuation index threshold, an alarm message indicating an abnormal cross-cluster backhaul algorithm is generated.

[0150] In an optional implementation, the device further includes: a recording module (not shown in the figure), configured to record the second node information of the originating node of the origin request corresponding to each origin record after generating the origin record corresponding to the origin request;

[0151] The acquisition module 520 is used to: after acquiring the back-to-source record generated within the preset time window, merge multiple back-to-source records corresponding to the same second node information into one back-to-source record;

[0152] And / or, an alarm module, for generating an alarm message indicating that a node corresponding to the second node information is abnormal if the number of source records corresponding to the same second node information exceeds a preset record number threshold.

[0153] It can be seen that in the source return exception processing device provided in the embodiment of the present application, after initiating a source return request, a source return record containing relevant information of the cluster where the initiating node is located, the requested resources, and the next-hop node is generated, so that according to the source return record, it is possible to accurately detect whether an abnormal repeated source return occurs within the preset time window, thereby achieving rapid and accurate detection of abnormal repeated source returns, improving the efficiency and accuracy of source return anomaly detection, saving detection costs, and being able to adapt to different protocols.

[0154] Embodiment 6

[0155] Figure 6 A structural schematic diagram of a computing device provided in Example 6 of the present application is shown. The specific embodiments of the present application do not limit the specific implementation of the computing device.

[0156] like Figure 6 As shown, the computing device may include: a processor (processor) 602 , a communications interface (Communications Interface) 604 , a memory (memory) 606 , and a communication bus 608 .

[0157] The processor 602, the communication interface 604, and the memory 606 communicate with each other via the communication bus 608. The communication interface 604 is used to communicate with other devices such as a client or other server network elements. The processor 602 is used to execute the program 610, which can specifically execute the relevant steps in the above-mentioned method for handling the back-to-source exception of the computing device.

[0158] Specifically, the program 610 may include program codes, which include computer operation instructions.

[0159] The processor 602 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0160] The memory 606 is used to store the program 610. The memory 606 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory. The program 610 may be specifically used to enable the processor 602 to perform the operations in the above method embodiment.

[0161] Embodiment 7

[0162] Embodiment 7 of the present application provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction or computer program, which enables a processor to perform operations corresponding to the source return exception handling method in any of the above method embodiments.

[0163] Embodiment 8

[0164] Embodiment 8 of the present application provides a computer program product, which includes at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to perform operations corresponding to the source return exception handling method in any of the above method embodiments.

[0165] To summarize, according to the computing device, computer storage medium, and computer program product provided in this embodiment, after initiating a back-to-source request, a back-to-source record containing relevant information of the cluster where the initiating node is located, the requested resources, and the next-hop node is generated, so that according to the back-to-source record, it is possible to accurately detect whether an abnormal repeated back-to-source occurs within a preset time window, thereby achieving rapid and accurate detection of abnormal repeated back-to-source, improving the efficiency and accuracy of back-to-source anomaly detection, saving detection costs, and being adaptable to different protocols.

[0166] The algorithm or display provided here are not inherently related to any specific computer, virtual system or other equipment. Various general systems can also be used together with the teaching based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the present application embodiment is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present application described here, and the above description of specific languages ​​is to disclose the best mode of implementation of the present application.

[0167] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0168] Similarly, it should be understood that in order to streamline the present application and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the embodiments of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following intention: the claimed application requires more features than the features clearly stated in each claim. More specifically, as reflected in the claims below, the inventive aspects are less than all the features of the single embodiment disclosed above. Therefore, the claims following the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present application.

[0169] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0170] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present application and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0171] The various component embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components according to the embodiments of the present application. The application can also be implemented as a device or apparatus program (e.g., computer program and computer program product) for executing a part or all of the methods described herein. Such a program implementing the present application can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0172] It should be noted that the above embodiments illustrate the present application rather than limit the present application, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be constructed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The present application may be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim that lists several devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A method for handling back-to-source exceptions, characterized in that: include: After any node of the content distribution network initiates a back-to-source request, a back-to-source record corresponding to the back-to-source request is generated; wherein the back-to-source record includes cluster information of the cluster to which the node initiating the back-to-source request belongs, resource information of the resource corresponding to the back-to-source request, and first node information of the node processing the back-to-source request; Get the back-to-source records generated within the preset time window; Detecting repeated target back-to-source records from the back-to-source records; Determine whether there is abnormal repeated return to the source within the preset time window based on the detected target return to the source record.

2. The method according to claim 1, characterized in that The method further includes: determining the number of target back-to-source records and / or the occurrence frequency of any target back-to-source record; Then, determining whether there is abnormal repeated return to the source within the preset time window according to the detected target return to the source record includes: determining whether there is abnormal repeated return to the source within the preset time window according to the number of target return to the source records and / or the occurrence frequency of any target return to the source record.

3. The method according to claim 2, characterized in that The determining whether there is abnormal repeated return to the source within the preset time window according to the number of target return to the source records and / or the occurrence frequency of any target return to the source record includes: If the number of target back-to-source records exceeds a preset number threshold, it is determined that abnormal duplicate back-to-source exists within the preset time window; And / or, according to the occurrence frequency of each target back-to-source record, the occurrence frequency accumulation value of the target back-to-source record in the preset time window is calculated, and if the occurrence frequency accumulation value exceeds a preset accumulation threshold, it is determined that abnormal repeated back-to-source exists in the preset time window.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: If it is determined that there is abnormal repeated return to the source within the preset time window, an alarm message is generated.

5. The method according to claim 4, characterized in that The method further comprises: If it is determined that there is abnormal repeated return to the source within the preset time window; Extract cluster information from each target back-to-source record; For any extracted cluster information, determine the duplication index of the cluster information according to the occurrence frequency of the target back-to-source record corresponding to the cluster information; Generate matching alarm information based on the duplication indicators of each cluster information.

6. The method according to claim 5, characterized in that The generating of matching alarm information according to the duplication index of each cluster information includes: Add the repeated indicators of each cluster information to the repeated indicator set; Calculating a volatility index of the set of repeated indicators; If the fluctuation index is greater than the preset fluctuation index threshold, the target cluster information is determined from the cluster information according to the repetition index of each cluster information, and an alarm message is generated indicating that the back-to-source algorithm of the target cluster corresponding to the target cluster information is abnormal; If the fluctuation index is less than or equal to the preset fluctuation index threshold, an alarm message indicating an abnormal cross-cluster backhaul algorithm is generated.

7. The method according to any one of claims 1 to 6, characterized in that After generating the back-to-source record corresponding to the back-to-source request, the method further includes: recording the second node information of the back-to-source request initiating node corresponding to each back-to-source record; After acquiring the back-to-source record generated within the preset time window, the method further includes: merging multiple back-to-source records corresponding to the same second node information into one back-to-source record; And / or, if the number of source records corresponding to the same second node information exceeds a preset record number threshold, an alarm message indicating that the node corresponding to the second node information is abnormal is generated.

8. A device for handling back-to-source exceptions, characterized in that: include: A generation module, configured to generate a back-to-source record corresponding to a back-to-source request after any node of a content distribution network initiates a back-to-source request; wherein the back-to-source record includes cluster information of a cluster to which the node initiating the back-to-source request belongs, resource information of a resource corresponding to the back-to-source request, and first node information of a node processing the back-to-source request; The acquisition module is used to obtain the back-to-source records generated within a preset time window; A detection module, used to detect repeated target back-to-source records from the back-to-source records; The determination module is used to determine whether there is abnormal repeated return to the source within a preset time window according to the detected target return to the source record.

9. A computing device, characterized in that include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to the source return exception processing method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The storage medium stores at least one executable instruction, and the executable instruction enables the processor to execute an operation corresponding to the method for handling a back-to-source exception according to any one of claims 1 to 7.

11. A computer program product, characterized in that The method comprises at least one executable instruction, wherein the executable instruction enables a processor to execute an operation corresponding to the method for handling a back-to-source exception as described in any one of claims 1 to 7.

Images